The Mom Test made its rounds on LinkedIn again today. A founder wrote that your mom is the worst person to validate your idea, not because she will lie to you, but because she loves you. Thousands of people nodded, saved the post, and will still run their next idea past the same three friends by Friday.
The advice is over a decade old and still loses to convenience every single week. It is worth being precise about why.
Why do friends always say it is a great idea?
Because you asked them a different question than the one you think you asked. You said, would you use this. They heard, do you believe in me. And the honest answer to the second question is yes, so that is the answer you get, wearing the costume of the first one.
Nobody is lying. Your friends are accurately reporting the state of the relationship. The mistake is filing that report under market research.
Is asking better questions enough to fix it?
It helps and it is not enough. The Mom Test's real rule, ask about past behavior instead of future intentions, removes the easiest way to be polite. What did you do the last time this happened is much harder to answer with encouragement than would you buy this.
But better questions do not fix a biased sample. A perfectly phrased question, put to someone whose main stake in the conversation is your feelings, still returns data about your feelings. On a Reddit thread this morning a founder asked people what idea they have been sitting on because they have no idea whether it is good. Sitting on an idea usually means every review so far came from people with no reason to be honest.
What are friends actually good for?
Plenty, just not verdicts. They can introduce you to people who genuinely have the problem. They can tell you when you are burning out. They can be early hands on a rough prototype, as long as you watch what they do with it and ignore most of what they say about it. A friend abandoning your app on day two is real data. A friend telling you it is great is weather.
What does honest validation look like instead?
Three properties, none of which a friendly conversation has. The person is a stranger, so the relationship is worth nothing and the answer has to stand on its own. The person actually has the problem, so their behavior around it already exists and can be asked about. And the signal costs something, time, money or effort, because signals that cost nothing are compliments.
A stranger who has never met you, describing what they already did about the problem last month, is worth a hundred dinner table endorsements. Not because strangers are smarter. Because they are the only people whose yes was not already decided before you finished the sentence.
Key takeaways
- Friends answer about the relationship, not the idea. Their yes is real, it just measures the wrong thing.
- Better questions reduce politeness but cannot fix a biased sample. Who you ask matters more than how you ask.
- Use friends for introductions and early hands, and read their behavior, not their opinions.
- Honest validation needs strangers who have the problem and a signal that costs the person something.
The people who love you are wonderful. That is exactly what disqualifies them.